maropu commented on a change in pull request #28123: URL: https://github.com/apache/spark/pull/28123#discussion_r432206204
########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/bucketing/CoalesceBucketsInSortMergeJoin.scala ########## @@ -0,0 +1,111 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.spark.sql.execution.bucketing + +import org.apache.spark.sql.catalyst.catalog.BucketSpec +import org.apache.spark.sql.catalyst.rules.Rule +import org.apache.spark.sql.execution.{FileSourceScanExec, FilterExec, ProjectExec, SparkPlan} +import org.apache.spark.sql.execution.joins.SortMergeJoinExec +import org.apache.spark.sql.internal.SQLConf + +/** + * This rule coalesces one side of the `SortMergeJoin` if the following conditions are met: + * - Two bucketed tables are joined. + * - The larger bucket number is divisible by the smaller bucket number. + * - "spark.sql.bucketing.coalesceBucketsInJoin.enabled" is set to true. Review comment: nit: could you use `COALESCE_BUCKETS_IN_JOIN_ENABLED` instead of `spark.sql.bucketing.coalesceBucketsInJoin.enabled`? ########## File path: sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala ########## @@ -2586,6 +2586,26 @@ object SQLConf { .checkValue(_ > 0, "The timeout value must be positive") .createWithDefault(10L) + val COALESCE_BUCKETS_IN_JOIN_ENABLED = + buildConf("spark.sql.bucketing.coalesceBucketsInJoin.enabled") + .doc("When true, if two bucketed tables with the different number of buckets are joined, " + + "the side with a bigger number of buckets will be coalesced to have the same number " + + "of buckets as the other side. This bucket coalescing can happen only when the bigger " + + "number of buckets is divisible by the smaller number of buckets.") Review comment: Could you describe that this rule is only applied to sort-merge joins? ########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/bucketing/CoalesceBucketsInSortMergeJoin.scala ########## @@ -0,0 +1,111 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.spark.sql.execution.bucketing + +import org.apache.spark.sql.catalyst.catalog.BucketSpec +import org.apache.spark.sql.catalyst.rules.Rule +import org.apache.spark.sql.execution.{FileSourceScanExec, FilterExec, ProjectExec, SparkPlan} +import org.apache.spark.sql.execution.joins.SortMergeJoinExec +import org.apache.spark.sql.internal.SQLConf + +/** + * This rule coalesces one side of the `SortMergeJoin` if the following conditions are met: + * - Two bucketed tables are joined. + * - The larger bucket number is divisible by the smaller bucket number. + * - "spark.sql.bucketing.coalesceBucketsInJoin.enabled" is set to true. + * - The difference in the number of buckets is less than the value set in + * "spark.sql.bucketing.coalesceBucketsInJoin.maxNumBucketsDiff". + */ +case class CoalesceBucketsInSortMergeJoin(conf: SQLConf) extends Rule[SparkPlan] { Review comment: nit: unnecessary space found: `extends Rule[SparkPlan] {` ########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/bucketing/CoalesceBucketsInSortMergeJoin.scala ########## @@ -0,0 +1,111 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.spark.sql.execution.bucketing + +import org.apache.spark.sql.catalyst.catalog.BucketSpec +import org.apache.spark.sql.catalyst.rules.Rule +import org.apache.spark.sql.execution.{FileSourceScanExec, FilterExec, ProjectExec, SparkPlan} +import org.apache.spark.sql.execution.joins.SortMergeJoinExec +import org.apache.spark.sql.internal.SQLConf + +/** + * This rule coalesces one side of the `SortMergeJoin` if the following conditions are met: + * - Two bucketed tables are joined. + * - The larger bucket number is divisible by the smaller bucket number. + * - "spark.sql.bucketing.coalesceBucketsInJoin.enabled" is set to true. + * - The difference in the number of buckets is less than the value set in + * "spark.sql.bucketing.coalesceBucketsInJoin.maxNumBucketsDiff". Review comment: ditto. ########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/bucketing/CoalesceBucketsInSortMergeJoin.scala ########## @@ -0,0 +1,111 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.spark.sql.execution.bucketing + +import org.apache.spark.sql.catalyst.catalog.BucketSpec +import org.apache.spark.sql.catalyst.rules.Rule +import org.apache.spark.sql.execution.{FileSourceScanExec, FilterExec, ProjectExec, SparkPlan} +import org.apache.spark.sql.execution.joins.SortMergeJoinExec +import org.apache.spark.sql.internal.SQLConf + +/** + * This rule coalesces one side of the `SortMergeJoin` if the following conditions are met: + * - Two bucketed tables are joined. + * - The larger bucket number is divisible by the smaller bucket number. + * - "spark.sql.bucketing.coalesceBucketsInJoin.enabled" is set to true. + * - The difference in the number of buckets is less than the value set in + * "spark.sql.bucketing.coalesceBucketsInJoin.maxNumBucketsDiff". + */ +case class CoalesceBucketsInSortMergeJoin(conf: SQLConf) extends Rule[SparkPlan] { + private def isPlanEligible(plan: SparkPlan): Boolean = { + def forall(plan: SparkPlan)(p: SparkPlan => Boolean): Boolean = { + p(plan) && plan.children.forall(forall(_)(p)) + } + + forall(plan) { + case _: FilterExec | _: ProjectExec | _: FileSourceScanExec => true + case _ => false + } + } + + private def getBucketSpec(plan: SparkPlan): Option[BucketSpec] = { Review comment: Could you move `getBucketSpec` and `isPlanEligible ` into `ExtractSortMergeJoinWithBuckets`? ########## File path: sql/core/src/test/scala/org/apache/spark/sql/execution/bucketing/CoalesceBucketsInSortMergeJoinSuite.scala ########## @@ -0,0 +1,141 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.spark.sql.execution.bucketing + +import org.apache.spark.sql.catalyst.catalog.BucketSpec +import org.apache.spark.sql.catalyst.plans.Inner +import org.apache.spark.sql.execution.{BinaryExecNode, FileSourceScanExec, SparkPlan} +import org.apache.spark.sql.execution.datasources.{HadoopFsRelation, InMemoryFileIndex} +import org.apache.spark.sql.execution.datasources.parquet.ParquetFileFormat +import org.apache.spark.sql.execution.joins.{BroadcastHashJoinExec, BuildLeft, SortMergeJoinExec} +import org.apache.spark.sql.internal.SQLConf +import org.apache.spark.sql.test.{SharedSparkSession, SQLTestUtils} +import org.apache.spark.sql.types.{IntegerType, StructField, StructType} + +class CoalesceBucketsInSortMergeJoinSuite extends SQLTestUtils with SharedSparkSession { + private def newRelation(numBuckets: Int): HadoopFsRelation = HadoopFsRelation( + location = new InMemoryFileIndex(spark, Nil, Map.empty, None), + partitionSchema = StructType(Seq(StructField("a", IntegerType))), + dataSchema = StructType(Seq(StructField("a", IntegerType))), + bucketSpec = Some(BucketSpec(numBuckets, Seq("a"), Seq("a"))), + fileFormat = new ParquetFileFormat(), + options = Map.empty)(spark) + + private def run( + leftNumBuckets: Int, + rightNumBuckets: Int, + isSortMergeJoin: Boolean, + expectedLeftCoalescedNumBuckets: Option[Int], + expectedRightCoalescedNumBuckets: Option[Int]): Unit = { + val leftRelation = newRelation(leftNumBuckets) + val rightRelation = newRelation(rightNumBuckets) + val left = FileSourceScanExec( + leftRelation, Nil, leftRelation.dataSchema, Nil, None, None, Nil, None) + val right = FileSourceScanExec( + rightRelation, Nil, rightRelation.dataSchema, Nil, None, None, Nil, None) + val join = if (isSortMergeJoin) { + SortMergeJoinExec(Nil, Nil, Inner, None, left, right) + } else { + BroadcastHashJoinExec(Nil, Nil, Inner, BuildLeft, None, left, right) + } + + val plan = CoalesceBucketsInSortMergeJoin(spark.sessionState.conf)(join) + + def verify(expected: Option[Int], subPlan: SparkPlan): Unit = { + val coalesced = subPlan.collect { + case f: FileSourceScanExec if f.optionalNumCoalescedBuckets.nonEmpty => + f.optionalNumCoalescedBuckets.get + } + if (expected.isDefined) { + assert(coalesced.size == 1 && coalesced(0) == expected.get) + } else { + assert(coalesced.isEmpty) + } + } + + verify(expectedLeftCoalescedNumBuckets, plan.asInstanceOf[BinaryExecNode].left) + verify(expectedRightCoalescedNumBuckets, plan.asInstanceOf[BinaryExecNode].right) + } + + test("bucket coalescing - basic") { + withSQLConf(SQLConf.COALESCE_BUCKETS_IN_JOIN_ENABLED.key -> "true") { + run( + leftNumBuckets = 4, + rightNumBuckets = 8, Review comment: Could you run the opposite condition (left=8, right=4 in this case) in all the tests for test coverage (https://github.com/apache/spark/pull/28123/files#diff-771ed8520558b80b80cfd4f222b4dc21R99-R103)? ########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/bucketing/CoalesceBucketsInSortMergeJoin.scala ########## @@ -0,0 +1,111 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.spark.sql.execution.bucketing + +import org.apache.spark.sql.catalyst.catalog.BucketSpec +import org.apache.spark.sql.catalyst.rules.Rule +import org.apache.spark.sql.execution.{FileSourceScanExec, FilterExec, ProjectExec, SparkPlan} +import org.apache.spark.sql.execution.joins.SortMergeJoinExec +import org.apache.spark.sql.internal.SQLConf + +/** + * This rule coalesces one side of the `SortMergeJoin` if the following conditions are met: + * - Two bucketed tables are joined. + * - The larger bucket number is divisible by the smaller bucket number. + * - "spark.sql.bucketing.coalesceBucketsInJoin.enabled" is set to true. + * - The difference in the number of buckets is less than the value set in + * "spark.sql.bucketing.coalesceBucketsInJoin.maxNumBucketsDiff". + */ +case class CoalesceBucketsInSortMergeJoin(conf: SQLConf) extends Rule[SparkPlan] { + private def isPlanEligible(plan: SparkPlan): Boolean = { Review comment: How about `isPlanEligible` -> `isScanOperation`? ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
